Conventional spatial distributions, such as the negative binomial and Neyman-Scott, have been proposed as models for the clustered distribution of ore deposits. B. B. Mandelbrot has proposed that ore deposits are clustered at many different scales and that fractal distributions are appropriate. The apparent fractal characteristics of geologic controls on the locations of hydrothermal deposits (self-similar systems of fractures, circulation cells, and magmatic intrusions) support his proposal. Examination of the spatial distribution of 4775 hydrothermal precious-metal deposits in the Basin and Range, western United States, reveals fractal distributions acting within two scale ranges, together covering the entire range from 1 to 1000 km.This result gives additional credibility to concepts of regional controls on mineralization, such as metallogenic provinces and tectonostratigraphic terranes. The distribution of these ore deposits about a known occurrence follows a regular radial-density law. Within a range of radii (r) of 1 to 15 km, the probability-density function of these deposits is given by d = 0.46 r-1.17, and within a range of radii of 15 to 1000 km, by d = 0.08 r-0.83 where d is deposits per square kilometre.